It began as a war for Kyiv and Odesa, but it is now being decided along a 300-kilometer arc around Kramatorsk. Both sides have already lost the war they originally planned to fight, and now they are bargaining for the right to call defeat a victory.
In February 2022, when Vladimir Putin announced the start of the “special military operation,” he specifically mentioned Odesa. By August 2026, more than half of Russia’s military force in Ukraine is concentrated around two cities in the Donetsk region whose combined prewar population was roughly 250,000. Kramatorsk and Sloviansk have neither ports nor strategic industries capable of altering the balance of power in Europe. Yet they are the focus of a positional war now in its fourth year, a conflict into which, by rough estimates, hundreds of billions of dollars and an irreplaceable share of the demographic resources of two countries have been poured.
This contraction of objectives is the central fact of the war’s current phase. Everything else, the negotiating rounds, strikes on oil refineries, burning e-commerce warehouses, paralyzed airports, is merely a superstructure built on top of that fact. The logic of the conflict can only be understood by taking it seriously.
How the War for Ukraine Became a War for Two Points on the Map
Contemporary conflict studies describe protracted wars through the concept of a contested resource. In the Russia-Ukraine war, that resource is formally Ukrainian sovereignty, but in practice it is the measurable component of sovereignty: Kyiv’s ability to control its own territory and refuse to recognize another state’s control over it. Over four and a half years, both sides have discovered through experience the limits of what they can realistically achieve.
The Kremlin failed to install a subordinate government in Kyiv. The Russian military failed to destroy Ukraine’s military infrastructure, a task that proves extraordinarily difficult even under far more favorable conditions. According to available information, a six-month U.S.-Israeli air campaign against Iran, a country without dependable allies and with highly exposed territory, failed to produce a decisive outcome. Ukraine, much of whose logistical rear is effectively located beyond its borders and whose military facilities are dispersed and sheltered, proved an even more difficult target for an adversary possessing an arsenal of strike capabilities clearly smaller than that of the United States.
Kyiv, for its part, has acknowledged that returning to the 1991 borders by military means is impossible for the foreseeable future. Given the current balance of forces and prevailing tactics, the front does not permit operational breakthroughs: both armies are capable only of grinding their way through tactical defenses.
The negotiating process imposed on the parties by Donald Trump in 2025 crystallized this mutual sobering-up into a specific formula. Putin has since repeatedly referred to the “spirit of Anchorage,” meaning his August meeting with the U.S. president in Alaska. According to sources on both sides, the substance of the compromise came down to the following: the withdrawal of the Armed Forces of Ukraine from the Donetsk and Luhansk regions, demarcation in the Zaporizhzhia and Kherson regions along the existing front line, and an immediate ceasefire once the first condition was fulfilled. Transcripts obtained by Bloomberg of conversations between Putin’s representatives, Kirill Dmitriev and Yuri Ushakov, and their American counterparts describe the same arrangement: the transfer of Donbas and an exchange involving “some other territories.”
The Kremlin has never officially confirmed this, for an obvious reason. Four regions have been incorporated into the Russian Constitution as constituent entities of the federation within their administrative borders. Agreeing to divide the Zaporizhzhia and Kherson regions along the front line would amount to a public abandonment of a constitutional provision, effectively acknowledging that the “new regions” are bargaining chips rather than sacred territory.
The Ukrainian side has repeatedly stated that the pressure is focused specifically on the Donbas issue, while the remaining questions, including guarantees against future aggression, have been presented as nearly resolved. The wording of Volodymyr Zelensky’s refusal is crucial: withdrawing troops from Donbas would divide the country and trigger a political crisis. That, apparently, is precisely the outcome the entire package of demands is designed to produce.
This is where the central intrigue of the negotiating process lies. Territory is formally the subject of the bargaining, but the real issue is the internal cohesion of Ukrainian society. Putin does not need Kramatorsk as much as he needs a Ukrainian government that surrendered Kramatorsk. The first gives him a line on a map. The second creates a political fracture that, over time, could yield more than any territorial boundary.
The Anchorage That Never Was: Why the Theory That “the Kremlin Is Deceiving Trump” Does Not Survive Scrutiny
One widespread interpretation holds that Moscow is participating in negotiations purely as a diversion while actually preparing to continue the war until it captures Kyiv and Odesa. That interpretation contradicts the deployment of Russian resources, the most reliable indicator of actual intentions.
In early 2025, Russian forces captured the southern part of the Donetsk region, including Kurakhove and Velyka Novosilka, and gained a theoretical opportunity to advance toward Dnipro along the Donetsk-Zaporizhzhia highway against a weakened Ukrainian grouping. Dnipro is a city of one million, a logistics hub, an industrial center, and a genuine strategic objective. Instead, the Russian General Staff redeployed an entire army consisting of two divisions from the Kurakhove sector to Kramatorsk. Reserves from the Kherson and Sumy sectors were later sent there as well.
That decision says more than any public statement. A military command with limited reserves spent them on the most heavily fortified sector rather than the most promising one. The explanation can only be political: Putin does not need Dnipro. He needs a specific administrative boundary embedded in the negotiating formula. Donbas has ceased to be merely a military objective and has become a bookkeeping entry in the balance sheet of his political project.
This leads to a practical conclusion: the probability of the war ending rises sharply under one of two scenarios. Either Russia comes close to capturing the Kramatorsk-Sloviansk urban agglomeration, or the Ukrainian military convinces the Kremlin that such a capture is impossible or would cost an unacceptable price. Everything else, strikes against rear areas, sanctions packages, diplomatic demarches, serves as background. These factors may affect how quickly one of those two outcomes is reached, but they cannot substitute for them.
Twelve Kilometers a Year: The Arithmetic That Makes Both Sides Uneasy
The current phase of the offensive against Kramatorsk began in May 2026. As of early August, the dynamics look like this.
Russian forces carried out an unsuccessful assault on Lyman northeast of Sloviansk, but approached to within 13 kilometers of the city from the east, advancing 17 kilometers since the beginning of the year. Another grouping is seven to nine kilometers from the southeastern edge of Kramatorsk after advancing roughly 14 kilometers since January. Kostiantynivka, south of Kramatorsk, has been almost completely captured. Russian assault groups have been spotted on the outskirts of Oleksiievo-Druzhkivka along the highway to Kramatorsk, 13 kilometers from the city, after advancing about 10 kilometers since the start of the year. Near Dobropillia, Russian forces have advanced roughly 12 kilometers, and only along one southern sector where the 76th Guards Air Assault Division is operating.
A linear extrapolation would place Russian forces on the outskirts of both cities in early 2027, followed by attempted assaults in the months thereafter. That calculation, apparently, is what the Russian leadership is working with. But it is vulnerable for several reasons.
The first is a structural feature of current tactics. Every breakthrough requires a pause lasting several weeks: clearing captured territory, preparing supply routes, moving drone crews closer to the front, establishing new artillery positions in the rear, and once again suppressing enemy fire assets, including drone operators, artillery, and electronic warfare systems. All of this must be done under continuous pressure from enemy drones. There are no longer sectors of the front where one tactical operation can immediately follow another. Only a year ago, the Vostok Group of Forces in the Zaporizhzhia region was advancing as much as 10 kilometers per week for several consecutive weeks. Such rates have been unattainable in the Donbas sector since late 2024. The offensive has become a discontinuous sequence of small operations separated by pauses.
The second reason is the condition of the army. In 2024, the Russian command used the influx of recruits and the reactivation of equipment from Soviet-era storage depots to create divisions and even corps- and army-level formations from scratch. Today, the only available option is to expand brigades into divisions using reserve regiments that have already been formed. Judging by this trend, the pace of reinforcement now exceeds the rate of attrition from deaths, wounds, and missing personnel only slightly. An army that stops growing loses the ability to increase pressure simultaneously across multiple sectors of the 300-kilometer front stretching from Lyman to the border of the Zaporizhzhia region.
The third reason lies in political psychology. Putin is demonstrably sensitive to the pace of operations. A prolonged assault on a dense urban agglomeration dominated by multi-story buildings, where the defending side enjoys every advantage, is the last thing a politician who promised a rapid operation four and a half years ago would want.
A Call to the Colonel: What Putin Actually Knows About the Front
One episode is worth a dozen expert assessments. In early August, Putin directly contacted Colonel Abdulaziz Shikhabidov, commander of the 76th Guards Air Assault Division, which has spent six months fighting its way toward Dobropillia north of Pokrovsk. Media attention focused on a secondary detail: a year earlier, Dagestani social media accounts had announced that Shikhabidov had been killed. The substance of the conversation matters far more.
The commander in chief asked the divisional commander exactly which settlements his units had reached. The colonel gave credible answers: his statements about fighting in villages on the southern outskirts of Dobropillia are corroborated by video footage, including material published by the Ukrainian side. Putin remarked with visible irritation that other advancing formations were performing worse. The key exchange concerned whether the assigned deadlines were realistic, and the colonel answered that they were.
A conversation conducted over the head of the General Staff with a divisional-level commander is itself a diagnosis of the command system. According to available information, forces of the 51st Combined Arms Army reported a breakthrough directly into Dobropillia, while published video footage shows that they have remained roughly five kilometers from the city for several months. Ukrainian intelligence claims that Russian forces routinely fail to meet command deadlines.
The conclusion from this episode has two parts, both of which challenge common assumptions. First, the theory that Putin is completely misinformed is not supported by the evidence. The Kremlin monitors the battlefield in considerable detail, down to individual settlements, and clearly does not trust the reports submitted by certain formations. Second, the existence of a “vertical of deception” at army and corps level is indirectly confirmed by the very fact that such a call was necessary. A leader forced to personally verify battlefield reports does not have a functioning feedback system.
Putin’s public formula, that Russian forces are advancing not as quickly as they would like but are moving forward every day, reads in this context as a rationalization. He is prepared for “victory” to require more time and more resources than originally expected and is redefining success in advance in terms of the continuity of forward movement rather than deadlines.
150 Versus 100: Why Money Does Not Decide the War
Russia’s military spending has reached 8 percent of GDP, more than twice its prewar level. Ukraine spends more than 55 percent of GDP, with a significant portion covered by Western countries through direct military assistance and financial support for the state budget. In absolute terms, the gap is relatively small: very roughly, $150 billion versus $100 billion based on 2024 figures.
A spending advantage of one and a half to one does not produce a decisive advantage in a war of attrition. That is perhaps the conflict’s central economic lesson. The reason lies in the structure of modern warfare: the dominance of inexpensive drones, communications, and intelligence over costly platforms has devalued the Soviet legacy on which Russia’s quantitative superiority was built. Much of the prewar weapons inventory failed to deliver the expected effectiveness, often not because of poor quality, but because the nature of the war did not match the assumptions under which military commanders had prepared. It is the classic mistake of armies that trained for a different war.
Economist Oleg Itskhoki has described Russia’s war-financing model through the principle of optimization: the screws are always tightened only as much as necessary, but they are tightened evenly. The state gradually extracts resources from the population through inflation, higher taxes, and the removal of labor through military recruitment, while avoiding any single point at which pressure becomes unbearable enough to provoke political protest. The system constructed since the fall of 2022 is stable and capable of functioning for a long time.
The weakness of the model lies in the same logic that makes it sustainable. Minimal optimization permanently rules out a rapid victory. Every year of war becomes more expensive: the cost of recruiting each new contract soldier rises, the cost of servicing government debt increases, and the pool of people willing to risk their lives for relatively modest sums shrinks. Resources that can be extracted without attracting too much attention are finite by definition. The need to keep the extraction inconspicuous is itself the limiting factor.
The alternative would be a radical restructuring of the economy, with military spending multiplied several times over and the entire population drawn into the war effort. Putin could have chosen that path in 2022 or later, but he has consistently preferred gradualism. That choice reveals his own assessment of the balance between the benefits of pursuing the war to outright victory and the domestic risks such an effort would create.
Ukraine’s mobilization model suffers from mirror-image problems. To keep pace with Russia in the number of replacements, the Armed Forces of Ukraine must draw on a several-times-larger share of the population. The coercive nature of conscription creates abuse and violence, damaging public morale and provoking desertion even before recruits reach the front. At the same time, neither country is close to exhausting its mobilization reserve. In Ukraine, according to available estimates, several million potential conscripts remain theoretically available.
The answer to the common question of why there are still enough soldiers in the fifth year of the war is simple and unpleasant: the demographic resources of the two countries substantially exceed the rate at which those resources are being consumed, even at the current level of losses.
Douhet Loses Again: A Century-Old Illusion Kyiv Is Testing Once More
By 2026, Ukraine had acquired sufficient numbers of two classes of drones: long-range systems capable of striking targets hundreds or even thousands of kilometers behind the front, and medium-range systems designed for targets 20 to 150 kilometers deep. A campaign began that Kyiv has described in various ways: a “40-day operation to influence Russia,” “bringing the war back to where it came from,” and “long-range sanctions.”
The targets included Russia’s largest oil refineries, extending as far as Western Siberia, followed by warehouses belonging to the Wildberries online marketplace. Medium-range drones equipped with video cameras and machine-vision capabilities began hunting vehicles in occupied territories, especially along highways linking the Rostov region with Crimea. According to The Bell, Wildberries suffered losses of at least 100 billion rubles from the attacks, and the company is projected to remain unprofitable for years.
The campaign operates on two tracks: worsening Russia’s economic position and reducing state budget revenues, while also depriving Russian society of the ability to pretend that the war does not concern it. Conceptually, this aligns with the approach advocated by former Ukrainian Defense Minister Mykhailo Fedorov, who proposed supplementing economic pressure by more than doubling Russian personnel losses so that Moscow could no longer recruit contract soldiers faster than it was losing them. Under this model, Putin would face a choice: end the war or impose compulsory mobilization, with unpredictable political consequences.
It is telling that Fedorov’s conflict with Commander in Chief Oleksandr Syrskyi ended with both men leaving their posts. The Ukrainian system itself never reached a consensus on whether the theory actually works.
There are grounds for skepticism. The idea that strategic bombing can force an adversary out of a war dates back to the Italian general Giulio Douhet, who formulated it after World War I: gain air superiority, then strike not military targets but those whose destruction would break the state’s and society’s will to resist. A century of practice has produced consistently disappointing results. American military analysts studying their own failures reduced them to a recurring pattern.
Even overwhelming air superiority does not make bombers invulnerable. Air-defense resistance forces commanders to adjust tactics and target sets. Target selection repeatedly fails to correspond with broader strategic objectives. Identified targets are quickly exhausted, forcing planners to find new ones. A race begins between the pace of destruction and the pace of repair. When attacks on industry fail to produce political effects, target selection shifts toward inflicting psychological pressure on the population. Strategic aviation is increasingly redirected against tactical targets. Each time, commanders explain failure by arguing that politicians prevented them from striking the right targets with sufficient intensity.
Until the late 1990s, inadequate precision was widely considered the root of the problem. The mass introduction of precision-guided weapons did not change the picture. In Yugoslavia, the outcome was comparatively favorable, but it cannot be separated from other factors: by a widely accepted assessment, Slobodan Milosevic’s government capitulated because a NATO ground operation appeared inevitable, the country was internationally isolated, and it lacked allies. The Libyan campaign required a second prolonged wave of attacks against tactical targets, combined with direct air support for opposition forces. The 2026 campaign against Iran failed according to the same pattern.
Even the atomic bombings of Japan, the canonical example of decisive air power, are less clear-cut than they appear. Tokyo’s realization that it had reached a strategic dead end was reinforced by the Red Army’s rapid advance through Manchuria, the empire’s principal resource base, while the emperor used the bombings as an argument to persuade the military leadership to accept surrender while preserving the throne.
Particular attention should be paid to the study “When Does Aerial Bombing Work?,” which examines air campaigns from 1917 to 1999. Its authors found a statistically significant result: populations subjected to bombing either do not blame their own governments or redirect the blame toward the enemy, regardless of the form of government. Applied to Russia, this suggests that people affected by Ukrainian strikes are highly likely to become more, not less, supportive of continuing the war.
The most accurate description of Ukraine’s campaign is therefore not a turning point but an attempt to convert a newly acquired technological capability into negotiating leverage. An additional effect is already well understood: this domain of warfare follows an escalatory logic. A side whose opponent retains the ability to retaliate should expect not an end to attacks but an intensification of reciprocal strikes. Civilian casualties in Ukraine are once again rising after several years of decline from the 2022 peak. In Russia, they are at a historic high.
Wildberries as a Military Target: Where the Boundary Actually Lies
On paper, marketplace warehouses meet the criteria of an attractive target: weak protection, the ability to destroy them with several drones carrying small warheads, large quantities of highly combustible goods, and damage to the state budget through lost value-added tax revenue. Tatiana Kim’s company surpasses Ozon by every major metric, from market share to total warehouse space, which explains why it has become the higher-priority target.
Under the law of armed conflict, however, warehouse complexes belonging to civilian companies, whether Wildberries in Russia or Nova Poshta in Ukraine, are not lawful military objectives, even if body armor or military components such as drone grenade-release devices are stored there. Nor does the argument that a victim of aggression has the right to respond symmetrically apply here. Such actions are classified as reprisals and are prohibited by international conventions.
A violation of the principle of distinction is far from the gravest category of war crime, and virtually every army involved in major wars over the past 80 years has committed such violations to one degree or another. The practical issue is therefore not classification but enforcement. A war moving toward a stalemate without a clear victor will almost certainly produce fragmented jurisdictions: each side will prosecute, within its own legal system, those it considers criminals. Russians may be prosecuted by the International Criminal Court, whose jurisdiction is not accepted by many states, including Russia itself. Russia, meanwhile, will continue sentencing Ukrainians under terrorism statutes rather than for war crimes, which are harder to prove, especially when the alleged crimes did not occur.
A separate issue is the algorithmic selection of targets. Alex Karp, the chief executive officer of Palantir, has publicly stated that his company’s models are actively involved in selecting targets for the Ukrainian military. Ukrainian intelligence has clarified that this includes the creation of a target database inside Russia. Palantir performed similar work under a Pentagon contract before the war with Iran. Targeting is becoming a service provided by a private corporation based in a third country, a circumstance that international humanitarian law has yet to meaningfully address.
Why No One Strikes the Leadership Offices, and Why the Nuclear Arsenal Remains Silent
The common question about whether Moscow and Kyiv have an unspoken agreement not to strike each other’s decision-making centers has a more prosaic answer. There is probably no agreement. What exists is a mutual understanding of symmetrical capabilities. Assassination attempts against lower-ranking politicians and military officers occur regularly on both sides. A strike against the top leadership would trigger a spiral of uncontrolled escalation in which both sides could lose control of the process.
Israel and the United States legitimized such attacks by eliminating part of Iran’s political and military leadership, while simultaneously demonstrating their strategic futility. The state did not collapse. The war did not end.
A similar logic explains the nonuse of tactical nuclear weapons. Three constraints are at work. The first is the nuclear taboo, which Putin himself appears, judging by his behavior, to share. The second is the position of key international partners: according to Western media reports, the leaders of China and India communicated their views on the nuclear scenario directly to Moscow. The third is military uselessness. Achieving a strategic advantage with a tactical nuclear weapon against a dispersed infantry-and-drone defense is extremely difficult. Triggering escalation with the West and global ostracism, by contrast, would be almost guaranteed.
The same logic explains why Russia has failed to paralyze Ukrainian logistics. Strikes on rail infrastructure and bridges have been conducted repeatedly, with largely disappointing results. Railroad tracks, apart from rolling stock and electrical systems, can be repaired quickly. Bridges can also be restored unless spans are collapsed or structural supports are damaged, and achieving that is extremely difficult. The fate of the bascule bridge across the Dniester Estuary near Zatoka in the Odesa region is illustrative: since 2022, the relatively fragile structure has been struck by ballistic missiles, glide bombs, and dozens of drones in individual raids, yet only in early August 2026 did Russian reports claim that one span had finally been brought down. The roadway across the Dnipro Hydroelectric Station, damaged by a missile strike in 2024, was quickly restored to traffic.
The asymmetry in strike effectiveness is not explained solely by superior Ukrainian intelligence, although that advantage exists. Ukraine has lived under wartime conditions for years. It conceals military production, imports a substantial portion of its resources, and receives much of the rest as foreign assistance. There are simply fewer vulnerable targets on its territory. Russia, by contrast, attempts to maintain ordinary life: factories, airports, and refineries operate openly, allowing Kyiv to select the most painful and least protected targets. The Russian side, relying on missiles with powerful warheads but limited terminal accuracy, turns every miss into a humanitarian disaster rather than a military effect.
What All This Means Beyond Ukraine
The Anchorage formula, even if never implemented, creates a precedent whose significance extends far beyond the Russia-Ukraine war. It is a model in which the front line is converted into legal status through the mediation of a great power, while the victim of aggression is required to voluntarily surrender territory that the aggressor failed to capture by force.
For the post-Soviet space, where virtually every border has an alternative interpretation, the question of whether such a model is recognized as legitimate matters more than any declaration about territorial integrity. The experience of recent years has shown that only a status legally recognized by the parties and backed by their own capabilities proves durable, rather than one resting on external guarantees. Security guarantees for Ukraine remain unresolved for the same reason they have remained unresolved elsewhere: a guarantor unwilling to fight is not a guarantor.
Four Scenarios and One Uncomfortable Conclusion
The first scenario is that the Russian army reaches the outskirts of Kramatorsk and Sloviansk in 2027 and captures the urban agglomeration. The Kremlin gains the ability to declare its objectives achieved and begin bargaining for sanctions relief in exchange for a ceasefire. The probability of the war ending is highest under this scenario, provided Moscow actually stops.
The second scenario is that the defense holds, the offensive runs out of momentum, and both losses and the cost of every kilometer continue to rise. The Kremlin is then forced to reassess whether the objective is achievable. A major Ukrainian counterattack near Kramatorsk would also reduce the risk of the war continuing after Donbas, precisely because it would demonstrate the limits of what Russia can achieve.
The third scenario is that Donbas is captured but the war continues with the objective of destroying Ukrainian statehood. Despite its superficial appeal, this would be the worst scenario for Moscow. Ukraine would gain an even stronger incentive to resist, while its allies would have a reason to maintain support. International-relations theory, through the concept of the security dilemma, predicts the opposite reaction from the one Moscow would seek: the stronger Russia becomes, the more Western support for Kyiv increases. Risks for the Kremlin would accumulate simultaneously across every dimension, from the breakdown of its carefully calibrated economic pressure as debt-servicing costs rise, to a change in the U.S. administration a little more than two years from now, to the depletion of the volunteer recruitment pool and the resulting need for compulsory mobilization.
The fourth scenario is apparently the one Russian negotiating tactics are designed to produce: under external pressure, Kyiv agrees to withdraw its forces, and a domestic political crisis accomplishes what the Russian military could not. A voluntary surrender of Donbas would, however, sharply increase the risk that the war continues, because it would demonstrate that further concessions can be extracted.
The common denominator among all four scenarios is that none ends in victory in the classical sense.
The War Will Not End in Victory
Public debate about the war is structured around the question, “Who will win?” The question is wrong. Both sides have already lost the war they originally intended to fight. Russia lost its war to politically remake Ukraine. Ukraine lost its war to fully restore territorial integrity by military force. What remains is a struggle for the right to determine whose defeat will be called victory, and the price of that right is measured in hundreds of thousands of human lives and trillions of rubles and hryvnias.
Putin built an economic model capable of sustaining war almost indefinitely, and that same model has permanently ruled out a rapid victory. The principle of minimal optimization is both the source of the system’s resilience and its sentence: a country in which pressure is applied evenly and almost imperceptibly is incapable of making a dramatic leap, and without such a leap, Kramatorsk takes years to capture. Ukraine has acquired a technological instrument for pressuring the Russian rear, only to discover that it has entered a domain in which a century of empirical evidence argues against it: air campaigns by themselves do not win wars, but they reliably consolidate the targeted society around its government.
The real outcome will be determined neither on the battlefield nor at the negotiating table, but in the accounting ledgers of two political systems, at the moment when one of them can no longer bear the cost of refusing to lose. Russia’s model gives the Kremlin time but deprives it of the ability to win quickly. Ukraine’s model gives Kyiv higher troop quality and Western support but requires a disproportionately large share of the population. Both systems are capable of functioning for years more.
That leads to the most uncomfortable conclusion. The fate of a war that has claimed more lives than any European conflict since 1945 now depends not on grand strategic decisions but on the speed at which the Russian army can advance ten to fifteen kilometers a year across the Donetsk steppe. It is this arithmetic, not the “spirit of Anchorage,” that will determine the shape of peace in Eastern Europe for decades to come.